On 14 July 2010 19:50, rogerdpack <[email protected]> wrote: >> I think the problem is the font size: the characters look to be made >> of single-pixel lines, which tesseract just doesn't handle well (and >> neither does anything else I've ever used, for that matter). I think >> speckle detection is the cause of this, but that's just a hunch. >> >> The image looks to have been generated; if you can control generation, >> set a larger font size. > > Thank you for your response. Unfortunately my resolution can't be > increased since it is a static box size. If I manually cut up the > digits I am able to OCR them with gocr, though tesseract seg faults, > that's for another e-mail :)
I'm looking into some stuff that can be done to improve recognition on generated images, but don't hold your breath. -- <Leftmost> jimregan, that's because deep inside you, you are evil. <Leftmost> Also not-so-deep inside you. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en.
